📜 Language Model Collaboration for Relation Extraction from Classical Chinese Historical Documents

November 25, 2025 · View on GitHub

This repository contains the code and data accompanying the paper “Language Model Collaboration for Relation Extraction from Classical Chinese Historical Documents.” It provides implementations for the SpERT small model and a Small-Language-Model + Large-Language-Model (SLM+LLM) collaboration framework, along with evaluation scripts.

📂 Data

  • All datasets are located in the chisre/ directory.
  • test.json and chisre_test.json contain the same records but in different formats. ➤ Use the file required by each specific script.

🧭 Repository Structure

🔧 Code Layout

▶️ Main Scripts in SLCoLM

  • python model_collaboration.py 🤝 Generates relation triplets using SpERT predictions, relation schemas, and example-driven generation.
  • python eval.py 📝 Evaluates the generated outputs. Make sure the input file path and expected format match what the script requires.

💡 Usage Recommendations

  • Use Python ≥ 3.8.

  • Install dependencies via requirements.txt (if provided).

  • Before running scripts, double-check:

    • 📁 Data paths
    • 🧩 Model weight paths
    • ⚙️ Config files

  • If you encounter errors, inspect the top of the script for configurable variables or search for config or paths in the repo.

📖 Citation

If you use this repository in your research, please cite:

@article{Tang_Wang_Wang_2026,
  title={Language model collaboration for relation extraction from classical Chinese historical documents},
  volume={63},
  ISSN={0306-4573},
  DOI={10.1016/j.ipm.2025.104286},
  number={1},
  journal={Information Processing & Management},
  author={Tang, Xuemei and Wang, Linxu and Wang, Jun},
  year={2026},
  month=jan,
  pages={104286}
}

📬 Contact

If you have any questions, feel free to: